HiDiGen: Hierarchical Diffusion for B-Rep Generation with Explicit Topological Constraints

2026-04-03

AI Summary

This work addresses the challenge of jointly modeling discrete topology and continuous geometry in boundary representations (B-reps) with deep generative models. To this end, the authors propose a hierarchical diffusion-based generative framework that decouples B-rep synthesis into two stages: first constructing a topological skeleton encoding face-edge associations, and then progressively generating and refining surface patches and vertex positions through a multi-stage diffusion process. Throughout generation, edge-vertex adjacencies are dynamically maintained to ensure structural consistency. Built upon a Transformer architecture, the method integrates explicit topological constraints with a hierarchical geometric generation strategy, thereby preserving topological validity while significantly improving the geometric accuracy and diversity of synthesized CAD models.

Abstract

Boundary representation (B-rep) is the standard 3D modeling format in CAD systems, encoding both geometric primitives and topological connectivity. Despite its prevalence, deep generative modeling of valid B-rep structures remains challenging due to the intricate interplay between discrete topology and continuous geometry. In this paper, we propose HiDiGen, a hierarchical generation framework that decouples geometry modeling into two stages, each guided by explicitly modeled topological constraints. Specifically, our approach first establishes face-edge incidence relations to define a coherent topological scaffold, upon which face proxies and initial edge curves are generated. Subsequently, multiple Transformer-based diffusion modules are employed to refine the geometry by generating precise face surfaces and vertex positions, with edge-vertex adjacencies dynamically established and enforced to preserve structural consistency. This progressive geometry hierarchy enables the generation of more novel and diverse shapes, while two-stage topological modeling ensures high validity. Experimental results show that HiDiGen achieves strong performance, generating novel, diverse, and topologically sound CAD models.

Problem

• B-rep generation
• topological constraints
• 3D CAD modeling
• geometric validity
• hierarchical generation

Innovation

• Hierarchical Diffusion
• B-Rep Generation
• Topological Constraints
• Transformer-based Diffusion
• CAD Modeling